- Issue created by @quietone
- last update
over 1 year ago 30,168 pass - @quietone opened merge request.
- First commit to issue fork.
- last update
over 1 year ago 30,168 pass - Status changed to Needs review
over 1 year ago 7:34am 20 September 2023 - 🇮🇳India atul4drupal
The changes look ok go go ahead with.
The MR pipeline was giving error most likely because it needed to be rebased, rebased and now it all looks ok (didnt realize that rebasing will add a comment in this thread too as a new MR)
RTBC +1 from my side, cant do it myself as I got 1 MR added. - Status changed to RTBC
over 1 year ago 1:39pm 20 September 2023 - 🇺🇸United States smustgrave
Change is simple enough. Also think this would of been a good one to tag for new users.
- last update
over 1 year ago 30,168 pass - last update
over 1 year ago 30,205 pass - last update
over 1 year ago 30,363 pass - last update
over 1 year ago 30,365 pass - last update
over 1 year ago 30,360 pass - 🇺🇸United States xjm
Hmm, #3283794: Fix 'should return {type} but return statement is missing' PHPStan L0 errors in test code → is interesting and seems a lot could have been the wrong way around. I think it's more common to have incorrect return docs than incorrect return values.
The diffstat for the previous issue (excluding the baseline regeneration) was
+192, -85
-- 277 total lines to review, so already above our recommended 100-200 and in the range that should be reserved for change sets that are scannable without context. That issue definitely is not scannable without context because you need to look up whole methods, callers, parent implementations, etc. to determine whether adding a return type or removing the docs is the correct approach, and what the type should be.For example, for this issue, I had to do:
grep -rl "recursiveKSort" *
...then open all those files and check how the method was used to see if they relied on the return value or the pass-by-reference. In all cases they relied on the pass-by-reference and did not expect a return value.
So this issue is fine, and committable by itself. However, can we first get a followup to check over #3283794: Fix 'should return {type} but return statement is missing' PHPStan L0 errors in test code → and ensure we didn't introduce any other incorrect return values?
And in the future, such issues should be split into smaller logical scopes.
- last update
over 1 year ago 30,361 pass - last update
over 1 year ago 30,379 pass - last update
over 1 year ago 30,377 pass - last update
over 1 year ago 30,382 pass - 🇳🇿New Zealand quietone
Follow up made, 📌 Check returns values added in #3283794 Active
- last update
over 1 year ago 30,392 pass - last update
over 1 year ago 30,394 pass - last update
over 1 year ago 30,397 pass - Status changed to Fixed
over 1 year ago 10:46am 13 October 2023 - 🇬🇧United Kingdom longwave UK
Committed and pushed 321492b8a8 to 11.x and 9164d2f354 to 10.2.x and 946c240fe7 to 10.1.x. Thanks!
-
longwave →
committed 946c240f on 10.1.x
Issue #3388204 by quietone, atul4drupal, xjm: Fix return type in \Drupal...
-
longwave →
committed 946c240f on 10.1.x
-
longwave →
committed 9164d2f3 on 10.2.x
Issue #3388204 by quietone, atul4drupal, xjm: Fix return type in \Drupal...
-
longwave →
committed 9164d2f3 on 10.2.x
-
longwave →
committed 321492b8 on 11.x
Issue #3388204 by quietone, atul4drupal, xjm: Fix return type in \Drupal...
-
longwave →
committed 321492b8 on 11.x
Automatically closed - issue fixed for 2 weeks with no activity.